This project proposes a combined theoretical and experimental approach to study important problems related to DNA-protein interactions. The major focus of the project is on the problem of nucleosome positioning along genomic DNA. The distributions of nucleosomes along eukaryotic genomes play roles in transcriptional regulation, recombination and repair of the double helix, so it is very important to understand the mechanism of nucleosome positioning. Solving the problem will bring better understanding of gene regulation and repair, two key elements in cancer development. We will determine a large set of parameters which specify sequence dependence of the conformational properties of DNA. The determination will be based on cyclization of short DNA fragments with specially designed sequences. This experimental method allows extracting DNA parameters with a unique precision, which is necessary for successful solution of the problem. We will carefully test the model of sequence specificity and the determined set of parameters. Our approach to the nucleosome positioning is based on the general assumption that the free energy of DNA elastic deformation, associated with nucleosome core binding, specifies the sequence dependence of the binding affinity. Using our model and the determined set of parameters we will be able to estimate this free energy for any particular DNA sequence. We propose to extent our studies on the enzymatic processes that bring together DNA sites separated along the molecule contour. Such processes require formation of loops by DNA molecule, that appear in nearly all major processes of DNA functioning. We will address a basic question regarding loop formation: namely, is looping a diffusion-limited process? Three different systems, DNA transcription, site-specific recombination, and type II restriction enzymes will be used to study the problem. The results of this part of the project will be used to address the problem of topological selectivity of site-specific recombination. The recombinases play an important role in the development of many diseases related to genome rearrangements. Better knowledge of the mechanism of their action is very important for development of new, potentially more specific therapies. Combining computer simulations and experiments, we will try to understand the mechanism of high topological selectivity of DNA resolvases.